Munich Personal RePEc Archive
Foreign Capital, National Welfare and
Unemployment in a Fair Wage Model
Chaudhuri, Sarbajit and Banerjee, Dibyendu
University of Calcutta
30 August 2009
Online at https://mpra.ub.uni-muenchen.de/18005/
MPRA Paper No. 18005, posted 22 Oct 2009 09:53 UTC
1
Foreign Capital, National Welfare and Unemployment in a Fair Wage Model
Sarbajit Chaudhuri
Dept. of Economics, University of Calcutta, India.
E-mail: [email protected]
and
Dibyendu Banerjee
Dept. of Economics, Serampore College, Dt. Hooghly, West Bengal, India.
E-mail: [email protected]
Address for communication: Dr. Sarbajit Chaudhuri, 23 Dr. P.N. Guha Road,
Belgharia, Kolkata 700083, India. Tel: 91-33-2557-5082 (O); Fax: 91-33-2844-1490 (P)
(This version: August 2009)
Abstract: The paper develops a three-sector general equilibrium model that can explain
simultaneous existence of unemployment of both skilled and unskilled labour. The
unemployment of unskilled labour is explicated in terms of rural-urban migration
mechanism while that of skilled labour is shown using the ‘fair wage hypothesis’. The
paper finds that foreign direct investment (FDI) in the primary export sector improve
both national welfare and urban unemployment problem of unskilled labour while the
consequences of foreign capital flows into the import-competing sector and high-skill
export sector are ambiguous. The paper justifies the desirability of FDI flow in the
primary export sector from the perspective of both unemployment and social welfare.
Keywords: Fair wage hypothesis; skilled labour; unskilled labour; national welfare;
unemployment.
JEL classification: F10, F13, J41, O15.
2
Foreign Capital, National Welfare and Unemployment in a Fair Wage Model
1. Introduction
There exists a voluminous theoretical literature that examines the welfare consequence of
foreign capital inflows in a small open economy both in the absence and presence of
unemployment. Brecher and Alejandro (1977) have considered a two-commodity, two-
factor full employment model while Khan (1982) has used a mobile capital Harris-
Todaro (hereafter HT) model with urban unemployment for the analytical purpose. The
important result, common to both which is well-known as the Brecher-Alejandro
proposition, is as follows. The inflow of foreign capital with full repatriation of its
earnings is necessarily immiserizing if the import-competing sector is capital-intensive
and is protected by a tariff. However, social welfare does not change in the absence of
any tariff. Here welfare is defined as a positive function of national income.
In the literature, the Brecher-Alejandro proposition has also been re-examined in terms of
three-sector models. The third sector may either be a duty-free zone (DFZ) (sometimes
called foreign enclave) as in the works of Beladi and Marjit (1992a, 1992b) or it may be
an urban informal sector as in the works of Grinols (1991) and Chandra and Khan (1993).
Beladi and Marjit (1992a) have shown that with full-repatriation of foreign capital
income, an inflow of foreign capital may lead to immiserizing growth in the presence of
tariff-distortion even if the foreign capital is employed in the export sector. This
generalizes the main result in the existing literature, which primarily focuses on foreign
capital inflow in the protected sector of the economy.
However, there are works like Marjit and Beladi (1996), Chaudhuri (2005, 2007) and
Chaudhuri et al. (2006) which have demonstrated how foreign capital might produce
favourable effect on welfare taking into consideration some essential features of the
developing economies e.g. labour market distortion, presence of vast informal economy
and non-traded goods. Chaudhuri (2007) and Chaudhuri et al. (2006) have adopted three-
3
sector HT framework that explains urban unemployment as a migration-equilibrium
phenomenon.
How to explain unemployment as a general equilibrium phenomenon depends on which
type of labour we are considering, skilled or unskilled. The Harris-Todaro (1970) type of
model is one way to explain unemployment in a general equilibrium setup where the
efficiency of each worker is considered to be exogenously given and equal to unity.
However, in such a model unemployment is specific to the urban sector and is suitable to
explain unemployment of unskilled labour only.1 But it does not account for the
unemployment of skilled labour which is a disquieting problem in a developing economy
particularly after global economic slowdown.
It is important to note that in an economy the possibility of being unemployed also rises
with increasing education and skills. In the case of India, NSSO surveys conducted over
the years show that the unemployment rate among those educated above the secondary
level was higher, in both rural and urban areas, than those with lesser educational
attainments. The NSSO 61st Round report, Employment and Unemployment Situation in
India 2004-05, attributes this to the fact that “the job seekers become gradually more and
more choosers as their educational level increases.” Serneels (2007) also has found that in
Ethiopia unemployment is concentrated among the relatively well-educated first time job
seekers who come from the middle classes.
For explaining the existence of unemployment of skilled labour one has to recourse to the
efficiency wage theories. A generalized version of efficiency wage theory is the ‘fair
wage hypothesis’ (FWH). Agell and Lundborg (1992, 1995), Feher (1991), Akerlof and
1 The involuntary unemployment of unskilled labour can also be explained by using the ‘consumption efficiency hypothesis’ (CEH) of Leibenstein (1957), Bliss and Stern (1978), Dasgupta and Ray (1986) etc. where the nutritional efficiency of a worker depends positively on his consumption level. The CEH is the earliest version of the efficiency wage theory and is applicable to the poor unskilled workers who are at or slightly above their subsistence consumption level.
4
Yellen (1990), etc. have explained unemployment as a general equilibrium phenomenon
using the FWH. As per the Agell and Lundborg (1992, 1995) treatment of the FWH, the
efficiency of a worker is sensitive to the functional distribution of income. Consequently,
the return to capital, wage rates and the unemployment rates of different types of labour
appear as arguments in the efficiency function.
The objectives of the present paper are as follows. First, as the developing economies are
plagued by both skilled and unskilled unemployment it is extremely important to develop
an analytical framework that can be useful in analyzing the consequences of different
policies on the national welfare and unemployment of both types of labour. We develop a
three-sector specific-factor Harris-Todaro type general equilibrium model where the
FWH is valid. Secondly, we intend to examine the validity of the standard immiserizing
result of foreign capital inflows using this setup. Finally, the consequences of capital
inflows on the unemployment of both types labour have been studied. The paper finds
that although an inflow of foreign capital into the primary export sector unambiguously
improves social welfare foreign capital inflows into the other two sectors may be welfare-
worsening. Unemployment situation of skilled labour unequivocally improves in both
cases. Finally, while capital of type N definitely lowers the urban unemployment of
unskilled labour, an inflow of the other type of capital may fail to improve the situation
unless the centripetal force of an increase in the rural sector wage is stronger than the
centrifugal force resulting from an increase in the expected urban wage. Therefore the
paper justifies the desirability of FDI flow in the primary export sector from the
perspective of both unemployment and social welfare.
2. The Model
We consider a small open dual economy with three sectors: one rural and two urban.
There are two types of capital, capital of type N and capital of type K, and two types of
labour, skilled and unskilled. The rural sector produces an agricultural commodity using
both types of capital and unskilled labour. Capital of type N is interpreted as a composite
5
input2 that is broadly defined to include not only natural resource like land but also
durable capital equipments e.g. tractors, harvesters, weed cutters, pump sets for irrigation
purpose. FDI in N implies an increased supply of the durable agricultural capital
implements. On the other hand, capital of type K is used to purchase inputs like
fertilizers, pesticides, weedicides etc. The capital (of type K)-output ratio in sector 1,
1,Ka is assumed to be technologically given.3 Sector 2 is an urban sector that produces a
low-skill manufacturing good by means of capital of type K and unskilled labour. Finally,
sector 3, another urban sector, uses capital of type K and skilled labour to produce a high -
skill commodity. As sectors 2 and 3 produce non-agricultural commodities capital of type
N is specific to the rural sector (sector 1). Skilled labour is a specific input in sector 3.
Unskilled labour is imperfectly mobile between sectors 1 and 2 while capital of type K is
completely mobile among all the three sectors of the economy. Sector 2 faces an
imperfect unskilled labour market in the form of a unionized labour market where
unskilled workers receive a contractual wage, *W , while the unskilled wage rate in the
rural sector, ,W is market determined. The two unskilled wage rates are related by the
Harris-Todaro (1970) migration equilibrium condition where the expected urban wage
equals the rural wage rate and .* WW Hence, there is urban unemployment of unskilled
labour. On the other hand, we use the FWH to explain unemployment of skilled labour
and the efficiency function is similar to that in Agell and Lundborg (1992, 1995). This
function can be derived from the effort norm of the skilled workers which is sensitive to
the functional distribution of income and the skilled unemployment rate. This is the
optimal effort function of the utility maximizing skilled workers. Capital of either type
2 This composite input is called ‘land-capital’ in the works of Bardhan (1973), Chaudhuri (2007) and Chaudhuri and Yabuuchi (2008). 3 Although this is a simplifying assumption it is not completely without any basis. Agriculture requires inputs like fertilizers, pesticides, weedicides etc. which are to be used in recommended doses. Now if capital of type K is used to purchase those inputs, the capital (K type)-output ratio,
1,Ka becomes constant technologically. However, labour and capital of type N are substitutes and the production function displays the property of constant returns to scale in these two inputs. However, even if the capital(K type)-output ratio is not given technologically the results of the paper still hold under an additional sufficient condition incorporating the partial elasticities of substitution between capital of type K and other inputs in sector 1.
6
includes both domestic capital and foreign capital. Incomes from foreign capital are
completely repatriated. Sectors 1 and 3 are the two export sectors while sector 2 is the
import-competing sector and is protected by an import tariff. Sector 2 uses capital of
type K more intensively with respect to unskilled labour vis-à-vis sector 1. Production
functions exhibit constant returns to scale4 with positive and diminishing marginal
productivity to each factor. Commodity 3 is chosen as the numeraire.
The following symbols will be used for formal presentation of the model.
Kia amount of capital of type K required to produce 1 unit of output in the ith sector,
i 1,2,3;
Nia = amount of capital of type N required to produce 1 unit of output in sector 1;
Lia unskilled labour-output ratio in the ith sector, i 1,2;
3Sa skilled labour-output ratio in sector 3 (in efficiency unit);
iP exogenously given relative price of the i th commodity, i = 1,2;
t ad-valorem rate of tariff on the import of commodity 2;
iX level of output of the i th sector, i 1,2,3;
E efficiency of each skilled worker;
SW wage rate of skilled labour;
SWE
wage rate per efficiency unit of skilled labour;
*W unionized unskilled wage in sector 2;
W competitive wage rate of unskilled labour in sector 1;
r return to capital of type K (both domestic and foreign);
R return to capital of type N (both domestic and foreign)
L endowment of unskilled labour (in physical unit);
S endowment of skilled labour (in physical unit);
v unemployment rate of skilled labour;
4 See footnote 3 in this context.
7
UL = urban unemployment of unskilled labour;
K economy’s aggregate capital stock of K type (domestic plus foreign);
N economy’s aggregate capital stock of N type (domestic plus foreign);
ji distributive share of the j th input in the i th sector for j ,N , ,L S K and i 1, 2,
3;
ji proportion of the j th input employed in the i th sector for j KL, and i 1,2,3;
'' proportionate change.
Given the perfectly competitive commodity markets the three price-unit cost equality
conditions relating to the three industries are as follows.
1 1 1 1L K NWa ra Ra P (1)
*2 2 2 (1 )L KW a ra P t (2)
3 3 1SS K
W a raE
(3)
Following Agell and Lundborg (1992, 1995) we assume that the effort norms of the
skilled labour depend positively on (i) skilled wage relative to average unskilled wage;
(ii) skilled wage relative to returns on capital of both types; and, positively on (iii) the
unemployment rate of skilled labour. It may be mentioned that the average unskilled
wage in the economy is the rural sector wage that follows from the ‘envelope property’ of
the HT framework.5 Therefore, we write
( , , , )S S SW W WE E vW r R
(4)
The efficiency function satisfies the following mathematical restrictions:
5 Unskilled workers are employed in the rural and low-skill urban manufacturing sectors where they earn W and *W wages, respectively. Some of the unskilled workers remain unemployed in the urban sector earning nothing. The average wage income of all unskilled workers in the economy is the rural sector wage. This can be easily shown from equations (10) and (11). So, the efficiency function, given by equation (4), indirectly takes into account the unioni zed wage and the urban unemployment of unskilled labour as determinants.
8
1 2 3 4, , , 0E E E E ; 11 22 33, , 0E E E ; 12 13 14 23 24 34 0E E E E E E .6
The unit cost of skilled labour in sector 3, denoted , is given by
( )(.)SW
E (5)
Each firm in sector 3 minimizes its unit cost of skilled labour as given by (5). The first-
order condition of minimization is
1 2 3S S SW W WE E E E
W r R (6)
where: iE s are the partial derivatives of the efficiency function with respect to
( )SWW
, ( )SWr
and ( )SWR
, respectively. Equation (6) can be rewritten as
1 2 3 1 (6.1)
where i is the elasticity of the (.)E function with respect to its i th argument. This is the
modified Solow condition as obtained in Agell and Lundborg (1992, 1995).
Full utilization of N and K types of capital respectively entail
1 1Na X N (7)
1 1 2 2 3 3K K Ka X a X a X K (8)
There is unemployment of skilled labour in the economy and the rate of unemployment is
.v The skilled labour endowment equation is, therefore, given by
3 3 (1 )Sa X E v S (9)
In the migration equilibrium there exists urban unemployment of unskilled labour. The
unskilled labour endowment equation is given by
1 1 2 2L L Ua X a X L L (10) 6 Mathematical derivation of the efficiency function from the rational behavior of a representative skilled worker and explanations of the mathematical restrictions on the partial derivatives are available in Agell and Lundborg (1992, 1995).
9
In a Harris-Todaro framework the unskilled labour allocation mechanism is such that in
the labor market equilibrium, the rural wage rate, ,W equals the expected wage income in
the urban sector. Since the probability of finding a job in the urban low-skill sector is
2 2 2 2( / ( ))L L Ua X a X L the expected unskilled wage in the urban sector is
2 2 2 2( * / ( ))L L UW a X a X L . Therefore, the rural-urban migration equilibrium condition of
unskilled labour is expressed as
2 2 2 2( * / ( ))L L UW a X a X L W ,
or equivalently,
2 2 1 1( * / ) L LW W a X a X L (11)
Using (7) and (9) equations (11) and (8) can be rewritten as follows. *
12 2
1
( ) LL
N
aW a X N LW a
; and, (11.1)
312 2
1 3
(1 )[( ) { }]KKK
N S
a E v Sa N a X Ka a
(8.1)
In this general equilibrium model there are ten endogenous variables; namely, , , ,W r R
1 2, , , , ,S UW E v L X X and 3X and the same number of independent equations; namely, (1) –
(4), (6), (7), (8.1), (9), (10) and (11.1). The endogenous variables are determined as
follows. The system does not possess the decomposition property. r is found from (2)
as *W is given exogenously. , , ,SW R W v and 2X are simultaneously solved from equations
(1), (4), (6), (8.1) and (11.1). E is then found from (3). 1X and 3X are solved from
equations (7) and (9), respectively. Finally, UL is found from (10).
A close look at the price system reveals that given the value of R , sectors 1 and 2 can be
conceived to form a Heckscher-Ohlin subsystem (HOSS) as they use two common
inputs: unskilled labour and capital of type K. It is sensible to assume that sector 2 is
more capital-intensive than sector 1 in value sense with respect to unskilled labour. This
implies that 2 2 1 1( / * ) ( / )K L K La W a a Wa .
10
We measure welfare of the economy by national income at world prices, Y , and is given
by
2 2(1 )D D SY WL RN rK W v S tP X (12)
It is assumed that the foreign capital incomes of both types are fully repatriated. In
equation (12), WL and (1 )SW v S give the aggregate wage incomes of the unskilled and
skilled workers, respectively. DRN and DrK denote rental incomes from domestic capital
of types N and K . Finally, 22 XtP measures the cost of tariff protection of the import-
competing sector.
3. Comparative Statics
We are now going to analyze the consequences of inflows of foreign capital on national
welfare and unemployment of both skilled and unskilled labour. An inflow of foreign
capital into the primary export sector is captured by an increase in the endowment of N
type of capital. On the other hand, the endowment of K type of capital goes up when
foreign capital flows into the other two sectors including the tariff-protected import-
competing sector. The incomes on foreign capital are completely repatriated.
Differentiating equations (1), (4) (6), (11.1) and (8.1) the following expressions are
derived, respectively.7
1 1ˆ ˆ 0L NW R (13)
1 3 4ˆ ˆ ˆ 0W R v (14)
1 2 3 4ˆ ˆ ˆ ˆ 0SBW B R B W v (15)
*5 6 2 2 1
ˆ ˆ ˆ ˆL LB W B R X N (16)
1 11 1 2 2 3 4 1
ˆ ˆ ˆ ˆ ˆ ˆˆ( ) ( )K NL K NL K K S KS W S R X W B v K N (17)
where:
7 Note that 1Ka is technologically given. See footnote 2 in this context.
11
2111 ( ) 0SWEB
E W ; 233
2 ( ) 0SE WBE R
;
2 2 2 3311 223 ( ) ( ) ( ) 0S S SW W W EE EB
W E r E R E
; 34 ( ) 0
1K vB
v
; (18)
* 1 15 2 1[ ( )] 0L L LN NLB S S ; 1 1
6 1[ ( )] 0L LN NLB S S ; and,
*2 2
*( ) 0.L LWW
Here kjiS is the degree of substitution between factors in sector k . For example,
1 1
1
( )( )LLL
L
aWSa W
, 1 1
1
( )( )LLN
L
aRSa R
etc. 0k
jiS for j k ; & 0kjjS .
Arranging (13) – (17) in a matrix notation one obtains
1 1
1 3 4
1 2 3 4*
5 6 2 11 1
1 1 3 4 2 12
ˆ 00 0 0ˆ 00 0
0ˆ0ˆ0 0 ˆ
ˆ ˆˆ
L N
S
L L
K NL K NL K K K
W
RB B B WB B NvS S B K NX
(19)
The determinant to the coefficient matrix is * 1 * 1
4 3 1 2 6 1 2 1 2 5 1 2( ) ( )L K K L NL N K K L NLD B B S B S
(+) (+)
*2 4 3 3 4 + [ ]L K J B B H (20)
(–)(+)
where
1 2 3 1 1 1{ ( ) ( )}L NJ B B ;
1 3 1 1( )L NH . (21)
12
As the production structure is indecomposable an increase in capital stock of type N must
decrease its rate of return, R i.e. ˆ ˆ( / ) 0R N . Thus, solving (19) by Cramer’s rule it can be
easily proved8 that
0D (22)
For determining the signs of J and H we need to impose some restrictions on the relative
responsiveness of the (.)E , 1E and 3E functions with respect to their two arguments:
( )SWW
and ( )SWR
. The efficiency function, given by equation (4), is assumed to satisfy the
following two special properties.
Property A The responsiveness of (.)E with respect to SWR
is greater than that with
respect to SWW
such that 3 1
1 1
( ) ( )N L
.
Property B The algebraic value of the elasticity of 3E with respect to SWR
is not less
than that of 1E with respect to SWW
i.e. 33 11
3 1
( ) ( ).S SE W E WE R E W
The implications of the above two properties are as follows. Although the efficiency of
the skilled workers depends on the relative income distribution, they are expected to have
different attitudes towards the earnings of different factors of production. So changes in
incomes of different factors should affect the efficiency of skilled labour in different
degrees. It is reasonable to assume that the average unskilled wage is substantially lower
than the skilled wage. That is why the skilled workers are expected to have a soft feeling
towards their unskilled counterparts. On the contrary, they would feel to be deprived
significantly if the returns on both types of capital increase relative to the skilled wage
8 This has been shown in Appendix I.
13
which badly affect their work morale. It is reasonable, therefore, to assume that increases
in incomes of the capitalists cause a greater negative response among the skilled workers
and lower their efficiency than that resulting from an increase in the average unskilled
wage.
Properties (A) and (B) of the efficiency function together imply that 9
1 1 1 1
1 3 3 2
( ) ( ) ( ); and,L
N
BB
(23)
1 2 3 1 1 1 1 3 1 1{ ( ) ( )} 0; ( ) 0L N L NJ B B H
Differentiating (2) and (3) it is easy to show that
ˆ ˆSE W (24)
This leads to the following corollary.
Corollary 1: The efficiency of skilled labour, E , and the skilled wage rate, SW , always
change in the same direction and in the same proportion.
From (13) we can write
1
1
ˆ ˆ( )N
L
W R
(25)
This establishes the following corollary.
Corollary 2: W and R are negatively correlated.
Using (25) equation (14) can be rewritten as follows.
3 1 1 1
4 1
ˆ( )ˆ L N
L
Rv
(26)
Using (23) from (26) the following corollary is imminent.
Corollary 3: R and v are positively related.10
9 This has been proved in Appendix II.
14
Adding (14) and (15) and substituting for W from (13) we get
1 3 2 1 1 1
1 3
( ) ( )ˆ ˆ[ ]L NS
L
B BW RB
(27)
(–)
With the help of (23) from (27) the following corollary immediately follows.
Corollary 4: R and SW are negatively related.
Solving (19) by Cramer’s rule the following proposition can be easily established. 11
Proposition 1: Under assumptions A and B, an inflow of either type of capital leads to (i)
an increase in the rural unskilled wage (W ); (ii) a decrease in the return to capital of type
N; (iii) an increase in the skilled wage ( SW );.(iv) a fall in the unemployment rate of
skilled labour ( v ); and, (v) an expansion of sector 3. Furthermore, (vi) sector 1 expands
(contracts) while sector 2 contracts (expands) owing to inflows of capital of type N (type
K).
An inflow of foreign capital of type N into sector 1 lowers its return, R . This raises the
value of marginal product of unskilled labour and hence the rural unskilled wage, W .
This becomes clear if one looks at equation (1). A fall in R lowers the skilled
unemployment rate, v (corollary 3) and raises the skilled wage, SW (corollary 4) and hence
their efficiency, E (corollary 1). As employment of skilled labour rises in efficiency unit
(also in physical unit) sector 3 expands and draws capital from the other two sectors.
Consequently, the capital-intensive sector 2 contracts and the unskilled labour-intensive
sector 1 expands following a Rybczynski type effect.
On the other hand, an inflow of foreign capital of type K cannot change its return, r , as it
is determined from equation (2). It produces a Rybczynski effect in the HOSS. Sector 2 10 As the rural sector unskilled wage and the return on capital of type N are negatively related (corollary 2) there is a negative relationship between the average unskilled (rural) wage and skilled unemployment rate. 11 See Appendix III for mathematical derivations of the results.
15
expands while sector 1 contracts as the former is more capital -intensive than the latter. As
sector 1 contracts the demand for capital of type N falls. This lowers R which in turn
raises bothW (corollary 2) and SW (corollary 4) and hence E (corollary 1) and lowers the
skilled unemployment rate, v (corollary 3). As the employment of skilled labour rises both
in efficiency and physical units sector 3 expands.
Differentiating the national income expression (equation 12) the following proposition
can be proved.12
Proposition 2: An inflow of foreign capital of type N is unambiguously welfare-
improving13 while inflows of K type of capital may fail to boost up social welfare.
We can explain proposition 2 in the following fashion. In proposition 1 we find that an
inflow of foreign capital of either type raises the aggregate unskilled wage, aggregate
skilled employment and hence the aggregate skilled wage but lowers the domestic capital
income of N type. The domestic capital income of K type, however, remains unchanged.
It is easy to show that the increase in the aggregate unskilled wage income dominates
over the fall in the capital income of N type. Hence in either case the aggregate factor
income unambiguously rises. Besides, an inflow of capital of N type leads to a
contraction of the tariff protected import-competing sector. Hence the cost of protection
of the import-competing sector falls. Social welfare unequivocally improves in this case.
But in the case of K type capital the protected sector expands. Hence there is no
guarantee that it improves social welfare unless the positive aggregate factor income
effect is strong enough to dominate over the negative distortionary effect of tariff
protection of the import-competing sector.
12 This has been proved in Appendix IV. 13 An inflow of foreign capital of type N raises the economy’s aggregate stock of durable agricultural capital implements. Here foreign capital inflow takes place into the economy’s primary export sector. There are other papers in the literature like Beladi and Marjit (1992a, 1992b) that have examined the welfare consequence of foreign capital into a small open economy’s export sector. However, they have found the inflow of foreign capital to be immiserizing.
16
Subtraction of equation (10) from (11) yields
2 2*( )U L
W WL a XW
(28)
Totally differentiating equation (28) one can establish the final proposition of the
model.14
Proposition 3: An inflow of foreign capital of type N unambiguously improves the urban
unemployment problem of unskilled labour. On the contrary, inflows of type K capital
improve the unemployment situation of unskilled labour
if 1 11 1 1
1 2
)1 (( )( )).L N LLN NL
N L
S S
We explain proposition 3 in the following manner. In the migration equilibrium the
expected urban wage for a prospective unskilled rural migrant equals the actual unskilled
rural wage. An inflow of foreign capital of either type affects the migration equilibrium
in two ways. First, the low-skill urban manufacturing sector expands or contracts. This
leads to a change in the number of jobs available in this sector. The expected urban wage
for a prospective rural migrant, 2 2[ * /{1 ( / )}],U LW L a X changes as the probability of
getting a job in this sector changes for every unskilled worker. This is the centrifugal
force. If the expected urban wage rises (falls) the centrifugal force is positive
(negative).This paves the way for fresh migration (reverse migration) from the rural
(urban) to the urban (rural) sector. On the other hand, an inflow of foreign capital of
either type raises the rural unskilled wage (see proposition 1). This is the centripetal force
that prevents rural workers from migrating into the urban sector. Thus, there are clearly
two opposite effects working on determination of the size of the unemployed urban
unskilled workforce. In the case of an inflow of foreign capital of type N the low-skill
urban manufacturing sector contracts both in terms of output and employment. The
expected urban unskilled wage falls. So the centrifugal force is negative and drives the
unemployed urban workers to return to the rural sector. Thus, both the centripetal and the
centrifugal forces work in the same direction and cause the urban unemployment of 14 See Appendix V for mathematical proof of this proposition.
17
unskilled labour to decline. On the contrary, in the case of an inflow of foreign capital of
K type the low-skill urban sector expands and causes the expected urban unskilled wage
to rise. This leads to more migration from the rural sector to the urban sector. Therefore,
in this case the centrifugal and centripetal forces work in the opposite direction to each
other. If the latter effect outweighs the former, the level of unemployment falls. This
happens under the sufficient condition as mentioned in proposition 3.
4. Concluding remarks
This paper has developed a three-sector general equilibrium model that can explain
unemployment of both skilled and unskilled labour. The unemployment of unskilled
labour is explicated in terms the of rural-urban migration mechanism while that of skilled
labour is shown using the ‘fair wage hypothesis’. Apart from the two types of labour
there are two types of capital in this model. Capital of type N is specific to the primary
export sector (sector 1) while capital of type K is used in all the three sectors of the
economy. Consequences of foreign capital inflows of both types have been studied on
national welfare and unemployment of either type of labour. The paper finds that an
inflow of foreign capital into the primary export sector unambiguously improves social
welfare. On the contrary, although inflows of capital of type K unquestionably improve
the aggregate factor income it may fail to improve social welfare. The unemployment
situation of the skilled labour unequivocally improves in both the cases. Finally, while
capital of type N definitely lowers the urban unemployment of the unskilled labour, the
effect of an inflow of type K capital may fail to improve the situation unless the
centripetal force of an increase in the rural sector wage is stronger than the centrifugal
force resulting from an increase in the expected urban wage. The paper, therefore,
justifies the desirability of FDI flow in the primary export sector from the perspective of
both unemployment and social welfare. It is important to mention that after witnessing
China’s exemplary success in the agricultural front the developing economies like India
are of late toying with the idea of permitting foreign investment in agriculture.15 The
15 See Deshpande (2007) in this context.
18
analysis of the paper provides a theoretical foundation of such a move by the developing
nations.
References:
Agell, J. and Lundborg, P. (1995): ‘Fair wages in the open economy’, Economica 62,
325-351
Agell, J. and Lundborg, P. (1992): ‘Fair wages, involuntary unemployment and tax policy
in the simple general equilibrium model’, Journal of Public Economics 47, 299-320.
Akerlof, G. and Yellen, J. (1990): ‘The fair wage effort hypothesis and unemployment’,
Quarterly Journal of Economics 105, 255-284.
Bardhan, P.K. (1973): ‘A model of growth of capitalism in a dual agrari an economy’. In:
Bhagwati, J.M., Eckaus, R.S. (Eds.), Development and Planning: Essays in Honour of
Paul Rosenstein-Rodan. USA, MIT Press.
Beladi, H. and Marjit, S. (1992a): ‘Foreign capital and protectionism’, Canadian Journal
of Economics 25, 233-238.
Beladi, H. and Marjit, S. (1992b): ‘Foreign capital, unemployment and national welfare’,
Japan and the World Economy. 4, 311-317.
Bliss, C.J. and Stern, N,H. (1978): ‘Productivity, wages, and nutrition, Parts I and II’,
Journal of Development Economics 5.
Brecher, R.A. and Alejandro, C.F. Diaz (1977): ‘Tariffs, foreign capital and immiserizing
growth’, Journal of International Economics 7, 317-322.
Chandra,V. and Khan, M. A. (1993): ‘Foreign investment in the presence of an informal
sector’, Economica 60, 79-103.
Chaudhuri, S. and Yabuuchi, S. (2008): ‘Foreign capital and skilled-unskilled wage
inequality in a developing economy with non-traded goods. In S. Marjit and E. Yu (eds.),
Contemporary and Emerging Issues in Trade Theory and Policy, Emerald Group
Publishing Limited, UK.
Chaudhuri, S. (2007): ‘Foreign capital, welfare and unemployment in the presence of
agricultural dualism’, Japan and the World Economy 19, 149-165.
19
Chaudhuri, S. (2005): ‘Labour market distortion, technology transfer and gainful effects
of foreign capital’, The Manchester School 73(2), 214-227.
Chaudhuri, S., Yabuuchi S. and Mukhopadhyay U. (2006): ‘Inflow of foreign capital and
trade liberalization in a model with an informal sector and urban unemployment’, Pacific
Economic Review 11,87-103.
Dasgupta, P. and Ray, D. (1986): ‘Inequality as a determinant of malnutrition and
unemployment’ Economic Journal 96, 1011-1034.
Deshpande, R.S. (2007): ‘Our approach must be guarded and cautious’, The Financial
Express (16.07.2007). Also available at http://www.financialexpress.com/news/our-
approach-must-be-guarded-and-cautious/205151/0
Feher, E. (1991): ‘Fair wages and unemployment’, Dept. of Economics, University of
Technology, Vienna.
Grinols, E. L. (1991): ‘Unemployment and foreign capital: the relative opportunity cost
of domestic labour and welfare’, Economica 57, 107-121.
Harris, J.R. and Todaro, M. P. (1970): ‘Migration, unemployment and development: a
two-sector analysis’, American Economic Review 60, 126-142.
Khan, M. A. (1982): ‘Tariffs, foreign capital and immiserizing growth with urban
unemployment and specific factors of production’, Journal of Development Economics
10, 245-256.
Leibenstein, H. (1957): Economic Backwardness and Economic Growth, Wiley, New
York.
Marjit, S. and Beladi, H. (1996): ‘Protection and gainful effects of foreign capital’,
Economics Letters 53, 311-326.
Serneels, P.M. (2007): ’The nature of unemployment among young men in urban
Ethiopia’, Review of Development Economics 11(1), 170-186.
20
Appendix I:
Solving (19) by Cramer’s rule the following result is obtained. *
1 4 3
ˆ( )ˆ L
R BDN
(A.1)
(+)(–)
where: * 1 * 1
4 3 1 2 6 1 2 1 2 5 1 2( ) ( )L K K L NL N K K L NLD B B S B S
(+) (+)
*2 4 3 3 4 + [ ]L K J B B H (20)
(–)
1 2 3 1 1 1{ ( ) ( )}L NJ B B ;
1 3 1 1( )L NH ; and, (A.2)
*1 2 2 1
*( ) 0L K L KWW
(Note that * 0 as sector 1 is more unskilled labour-intensive vis-à-vis sector 2 in value
sense.)
In an indecomposable production structure like this it is sensible to assume that R falls
(rises) if N rises (falls) i.e.ˆ
( ) 0.ˆRN
From (A.1) it then follows that
0D (22)
From (20), (A.2) and (22) it follows that two sufficient conditions for 0D are:
, 0J H .
21
Appendix II:
As 1 3( ); ( ) 0( ) ( )S S
E EE EW WW R
and 11 33, 0E E we must have
21 11[ ( ) ] 0SWE E
W ; and,
23 33[ ( ) ] 0SWE E
R . Using (18) one can write
1 1( ) 0B ; and, (A.3)
3 2( ) 0B .
From Assumption A it follows that
1 1
1 3
( ) ( )L
N
(A.4)
That 0H is a direct consequence of Assumption A. We are going to prove that 0J if
Assumption B holds.
From (23)
0J 1 1 1
1 3 2
( ) ( )L
N
BB
(A.5)
Now
1 2 3 11 1 1 1 2 1
3 2 3 3 2 3 2 3 3 1
( )( )[ ] [ ] ( )[ ]( ) ( )
B BB B BB B B
Substituting the values of 1B and 2B from (18) and simplifying we can obtain the following
expression.
33 111 1 1 1
3 2 3 3 2 3 1
( )[ ] ( )[ ]( )
S SE W E WBB B E R E W
(A.6)
Now if 33 11
3 1
( ) ( )S SE W E WE R E W
i.e. if Assumption B holds from (A.3) and ( A.6) it follows that
22
1 1 1
3 2 3
( )( )BB
(A.7)
From (A.4) and (A.7) we can write
1 1 1
1 3 2
( ) ( )L
N
BB
0J .
Combining (A.4) and (A.7) and using (21) one can write
1 1 1 1
1 3 3 2
( ) ( ) ( ) , 0.L
N
B J HB
(22)
Appendix III:
Solving (19) by Cramer’s rule, using (18), (22) and (23) and simplifying the following
results can be obtained.
*4 1 3ˆ
0ˆN BW
DN
;
*4 1 3 2
ˆ0ˆ
N LBWDK
;
*
1 4 3
ˆ( ) 0ˆ L
R BDN
;
*2 1 4 3
ˆ ( ) 0ˆL L BR
DK
*
4
ˆ0ˆ
SW JDN
;
*2
4
ˆ0ˆ
S LW JDK
*
3ˆ
0ˆv B H
DN
;
*2
3ˆ
0ˆLv B HDK
(A.8)
* 1 *12 4 3 3 4 2 4 3 1 1 1 1 2
ˆ 1 [ ( ) { ( ) } ] 0ˆ D L K K L LN N L N LX J B B H B SN
1 *
3 2 411 1
ˆ( ) 0ˆ
NL LL N
B SXDK ; 3 4 1 6 1 52
ˆ ( ) 0ˆL NB B BX
DK
1 *21 4 3 3 4 1 4 3 1 1 1 1 2
ˆ 1 [ ( ) { ( ) }] 0ˆ L K K L LN N L N LX J B B H B S
DN
23
Results presented in (A.8) have been verbally stated in proposition 1.
Appendix IV:
Totally differentiating (12), using (A.8), (18), (22) and (23) and simplifying the following
two expressions can be derived * *
4 31 1 4 3
ˆ( ) ( ) {(1 ) }ˆ
SN L D
B W SYY WL RN v J vB HD DN
(+) (+) ( –)(+)
1 *2 21 3 4 4 3 1 4 3 1 1 1 1 2[ ( ) { ( ) }]L K K L LN N L N L
tP X B B H J B SD
(A.9)
(+) (–) (+) (+) (–) (+)
and,
*4 31 1 2
ˆ( ) ( )ˆ N L D L
BYY WL RNDK
*
24 3{(1 ) }S LW S v J vB H
D
(–) (+) (+) ( –)(+)
4 32 2 1 6 1 5{( ) ( )}L N
B tP X B BD
(A.10)
(–) (+)
Now
1 1 1 11
( ) ( ) 0DN L D N L
N
NWL RN W L aa
(A.11)
(as from (7) 11
D
N
N Xa
; and, DN N )
Using (A.11) from (A.9) we can conclude that
ˆ( ) 0ˆYN
.
However the sign of ˆ
( )ˆYK
is ambiguous which is clear from (A.10).
24
Appendix V:
Total differentials of equation (28) yield * *
2 2ˆ ˆ ˆ[( ) ( ) ]LU U L
W W WL X WW W
(A.12)
where ( )ULU
LL
Using (A.8) and simplifying from (A.12) the following expressions can be derived. *
21 3 4 3 4
ˆ( ) ( )[[( )[ ( )ˆ
U LL K
LU
L W W B B H JD WN
1 *
1 4 3 1 1 1 1 2{ ( ) }]K L LN L N N LB S
(+) ( –)(+)(+) (+)(+) (+)( –) (+)
*
*1 4 3( )( ) ]] 0.N
W BW
(A.13)
(–) (+)
1 14 3 21 1 1 1 2
ˆ *( ) ( )[( ) ( ) ( ){ˆU L
L N L LN NL N LLU
L B WS SD WK
1 11 1 1( ) ( )}]L N L LN NLS S
(A.14)
From (A.14) it follows that
ˆ( ) 0ˆ
ULK
if 1 11 1 1
1 2
)1 (( )( ))L N LLN NL
N L
S S
(A.15)